7 ms·

Why Our Memory Fails Us

76 points by kafkaesque 12y ago

ars 12y ago
Just a little thought experiment: What if those memories that differ between people are not because of bad memory, but because the actual events are different? Sort of a quantum superposition, which collapses the wavefront in different ways for different people? Maybe tie in multiple universes somehow, and events shift from one to the other. Would certainly be interesting if true :)
- alexbecker 12y ago
  > Would certainly be interesting if true :) Yes, except it's psuedoscientific crazy talk.
  - blueflow 12y ago
    Everything was "psuedoscientific [sic!] crazy talk" before it was widely recognized in the science world.
    - noonespecial 12y ago
      Thats a good example of something that is technically true but not useful. Like saying that all great men were once babies.
      - paulannesley 12y ago
        I don't think so. Those great men (and women) thought to once be babies were in fact once babies. Whereas those truths thought once to be pseudo-scientific crazy talk were actually truths all along. Thinking they were crazy talk was a mistake, and we can use that lesson to avoid repeating the mistake. Useful.
  - ars 12y ago
    Is that a bad thing? It's a thought experiment.
    - gizmo686 12y ago
      What is the point of this thought experiment? If I know I am looking for a certain thing in a thought experiment, than I might be able to suspend disbelief on the other parts, but without knowing what I am looking for, all I see are the inherit inconsistencies in the universe you posit.
      - ars 12y ago
        It doesn't really have a point. I was just trying to find a way to make it possible for people to actually remember different things and both to be correct. So like I remember one thing, but once the waveform collapsed it ended up being different. But my memories are either of what could have been (the other state of the waveform), or alternately memories from a different universe. I thought it was an interesting idea so I wrote it. I'm not trying to start new physics.
  - ChuckMcM 12y ago
    If by pseudo-scientific you mean difficult/impossible to falsify then I'd agree with you. If you mean gobbledy gook then I would disagree. There are some number of theorists that are working with multiple universe theories defined by 'choices' in the quantum world. Its hard to prove that such things exist (or don't exist) but one notion has that detangling a photon from its entangled counterpart has you "actualizing" in one of two possible universes. The GP thread posits macroscopic effects of that, which is both impossible to prove and an interesting conjecture.
    - gizmo686 12y ago
      The GP posits interactions between the several possible universes, in the sense that I am from universe A and am litterally talking to someone from universe B. While it is conceivable that this type of cross universe interaction is possible, it is not at all possible in our current understanding of physics, and if it where possible, it would have to be possible in a very subtle way or else we would have already noticed it (in the same sense that relativity breaks Newtons laws in a very subtle way). Assuming that such interactions our possible, we have no reason to believe that the macroscopic effects would be to transfer intact memories between universes, or for the universes to be similar enough for us not to notice this effect explicitly, but different enough for events to occur differently for different people.
      - ChuckMcM 12y ago
        Fair enough. I'm not an adherent to the many worlds interpretation, and I claim no specific expertise in this space. That said, I read your comment as implying that a MWI would be, by definition, acyclic? To be honest I was thinking that might not be the case on the theory that you could arrive at the same universe by many paths. Some part of me wants infinity constrained slightly :-)
      - ars 12y ago
        Only the memory is transfered from one universe to another, not the actual event. And because people don't usually falsely remember things in a way that is inconsistent with how things turned out only subtle changes would be transfered. This is just an idea I'm playing with for fun, I'm not trying to start new physics.
- Florin_Andrei 12y ago
  Okay, I'm gonna take the bait. > Sort of a quantum superposition, which collapses the wavefront in different ways for different people? Maybe tie in multiple universes somehow, and events shift from one to the other. Then the observers with differing memories would actually live in different "multiverse tracks" (different universes with a common origin in the past) and would never interact with each other, ever again. The fact that they can interact, even indirectly via a 3rd party like a cross-examiner, is proof that they live in the same worldtrack, and therefore the wave has "collapsed" the same way for both.
- 4ydx 12y ago
  So talking about something that happens on a very very small scale as if it should happen on a large one? You sound like Depak Chopra. But don't worry we will have something similar to this in the not so distant future when editing video footage quickly and with picture perfect results makes it hard to determine if evidence is real.
tokenadult 12y ago
Christopher Chabris is a very perceptive researcher on human perception, and I'm glad to see this submission here. The lead example of faulty memory in the article is very interesting, as it points to the problem of belief-motivated recollection of "facts" that never happened. This is important for all of us, as the article notes. "Our lack of appreciation for the fallibility of our own memories can lead to much bigger problems than a misattributed quote. Memory failures that resemble Dr. Tyson’s mash-up of distinct experiences have led to false convictions, and even death sentences. Whose memories we believe and whose we disbelieve influence how we interpret controversial public events, as demonstrated most recently by the events in Ferguson, Mo." As I have written about this issue before here on Hacker News, the problem with human memory is that each read operation is also a write operation. We never go back to our memory of any event without rearranging our understanding of the event, often to the detriment of recalling verifiable facts. When in doubt, look it up. And be in doubt often when relying on your own memory.
- api 12y ago
  I also like that he picks on Tyson... not because I don't like Tyson, but because it's important to point out that these are universal human cognitive failings that can befall anyone regardless of their belief system. Nobody who has a human brain is immune to the issues with that organ. Too often "Skeptic movement" people tend to pick on religious believers, new agers, etc. as if obviously only those kinds of people fall prey to cognitive biases, illusions, logical fallacies, and other failure modes of the mind.
  - DanBC 12y ago
    > Too often "Skeptic movement" people tend to pick on religious believers, new agers, etc. as if obviously only those kinds of people fall prey to cognitive biases, illusions, logical fallacies, and other failure modes of the mind. That was particularly annoying on the bad science forums. Especially since the whole point of the book was that people are subject to biases and well run research is the way to combat those biases.
- WalterBright 12y ago
spiritplumber 12y ago
Why Our Memory Fails Us: Because we don't have badges. Cops have magically perfect memory, even to the point that their memory trumps dashcam video. https://www.techdirt.com/articles/20140405/17142626817/indiana-supreme-court-declares-officers-testimony-is-more-reliable-than-video-evidence.shtml https://www.techdirt.com/articles/20140405/17142626817/india...
dewarrn1 12y ago
False memories of the kind that Chabris and Simons discuss are ubiquitous in healthy people, and there are lots of interesting wrinkles in the literature describing these memory errors — here are two of my favorites. First, even individuals with extraordinary memory abilities are susceptible to false memory phenomena [0]. People with what's been called "highly superior autobiographical memory" (HSAM) can, for any given date in the last decade or two, tell you what they did, what they ate, and what the weather was like with extraordinary accuracy. However, when tested with standard laboratory tasks that induce false memories, they are just as vulnerable as typical individuals. Second, focal brain injury may reduce susceptibility to false memories [1]. Full disclosure, this is some of my own published work. We investigated the effects of damage to the ventromedial prefrontal cortex (vmPFC) on false recall in a standard laboratory test involving lists of conceptually similar words (the DRM task [2]). Patients with vmPFC damage showed reduced (but not abolished) false memories relative to healthy comparison participants. This is especially interesting to us because vmPFC appears to be necessary for decision-making in complex situations, particularly when prior knowledge must be integrated into ongoing behavior. We believe that deficits in decision-making accompanying vmPFC damage may stem from the same root cause the reductions in false memory that we observed. Regardless, we’re not recommending vmPFC resection for healthy people seeking more accurate memory (not yet, anyway). Although these are highlights, empirical work on false memories and other kinds of memory errors extends back more than a century. It’s a fascinating domain, and I very much enjoy working in it. More information on a variety of related topics can be found at my website [3] and in my published work (see Publications). [0] https://www.ncbi.nlm.nih.gov/pubmed/24248358 https://www.ncbi.nlm.nih.gov/pubmed/24248358 [1] https://www.ncbi.nlm.nih.gov/pubmed/24872571 https://www.ncbi.nlm.nih.gov/pubmed/24872571 [2] https://en.wikipedia.org/wiki/Deese%E2%80%93Roediger%E2%80%93McDermott_paradigm https://en.wikipedia.org/wiki/Deese%E2%80%93Roediger%E2%80%9... [3] http://david-e-warren.me http://david-e-warren.me
colmvp 12y ago
The article brings up Dr. Tyson. He actually commented on this article and linked to these Facebook posts: https://www.facebook.com/notes/neil-degrasse-tyson/email-exchange-with-the-federalist/10152354422065869 https://www.facebook.com/notes/neil-degrasse-tyson/email-exc... https://www.facebook.com/notes/neil-degrasse-tyson/partial-anatomy-of-my-public-talks/10152360009440869 https://www.facebook.com/notes/neil-degrasse-tyson/partial-a...
guelo 12y ago
Disappointingly, this article does not tell us why our memory fails us, only that it does.

76 points by kafkaesque 12y ago

ars 12y ago

Just a little thought experiment: What if those memories that differ between people are not because of bad memory, but because the actual events are different? Sort of a quantum superposition, which collapses the wavefront in different ways for different people? Maybe tie in multiple universes somehow, and events shift from one to the other. Would certainly be interesting if true :)

alexbecker 12y ago
> Would certainly be interesting if true :) Yes, except it's psuedoscientific crazy talk.
- blueflow 12y ago
  Everything was "psuedoscientific [sic!] crazy talk" before it was widely recognized in the science world.
  - noonespecial 12y ago
    Thats a good example of something that is technically true but not useful. Like saying that all great men were once babies.
    - paulannesley 12y ago
      I don't think so. Those great men (and women) thought to once be babies were in fact once babies. Whereas those truths thought once to be pseudo-scientific crazy talk were actually truths all along. Thinking they were crazy talk was a mistake, and we can use that lesson to avoid repeating the mistake. Useful.
- ars 12y ago
  Is that a bad thing? It's a thought experiment.
  - gizmo686 12y ago
    What is the point of this thought experiment? If I know I am looking for a certain thing in a thought experiment, than I might be able to suspend disbelief on the other parts, but without knowing what I am looking for, all I see are the inherit inconsistencies in the universe you posit.
    - ars 12y ago
      It doesn't really have a point. I was just trying to find a way to make it possible for people to actually remember different things and both to be correct. So like I remember one thing, but once the waveform collapsed it ended up being different. But my memories are either of what could have been (the other state of the waveform), or alternately memories from a different universe. I thought it was an interesting idea so I wrote it. I'm not trying to start new physics.
- ChuckMcM 12y ago
  If by pseudo-scientific you mean difficult/impossible to falsify then I'd agree with you. If you mean gobbledy gook then I would disagree. There are some number of theorists that are working with multiple universe theories defined by 'choices' in the quantum world. Its hard to prove that such things exist (or don't exist) but one notion has that detangling a photon from its entangled counterpart has you "actualizing" in one of two possible universes. The GP thread posits macroscopic effects of that, which is both impossible to prove and an interesting conjecture.
  - gizmo686 12y ago
    The GP posits interactions between the several possible universes, in the sense that I am from universe A and am litterally talking to someone from universe B. While it is conceivable that this type of cross universe interaction is possible, it is not at all possible in our current understanding of physics, and if it where possible, it would have to be possible in a very subtle way or else we would have already noticed it (in the same sense that relativity breaks Newtons laws in a very subtle way). Assuming that such interactions our possible, we have no reason to believe that the macroscopic effects would be to transfer intact memories between universes, or for the universes to be similar enough for us not to notice this effect explicitly, but different enough for events to occur differently for different people.
    - ChuckMcM 12y ago
      Fair enough. I'm not an adherent to the many worlds interpretation, and I claim no specific expertise in this space. That said, I read your comment as implying that a MWI would be, by definition, acyclic? To be honest I was thinking that might not be the case on the theory that you could arrive at the same universe by many paths. Some part of me wants infinity constrained slightly :-)
    - ars 12y ago
      Only the memory is transfered from one universe to another, not the actual event. And because people don't usually falsely remember things in a way that is inconsistent with how things turned out only subtle changes would be transfered. This is just an idea I'm playing with for fun, I'm not trying to start new physics.
Florin_Andrei 12y ago
Okay, I'm gonna take the bait. > Sort of a quantum superposition, which collapses the wavefront in different ways for different people? Maybe tie in multiple universes somehow, and events shift from one to the other. Then the observers with differing memories would actually live in different "multiverse tracks" (different universes with a common origin in the past) and would never interact with each other, ever again. The fact that they can interact, even indirectly via a 3rd party like a cross-examiner, is proof that they live in the same worldtrack, and therefore the wave has "collapsed" the same way for both.
4ydx 12y ago
So talking about something that happens on a very very small scale as if it should happen on a large one? You sound like Depak Chopra. But don't worry we will have something similar to this in the not so distant future when editing video footage quickly and with picture perfect results makes it hard to determine if evidence is real.

tokenadult 12y ago

Christopher Chabris is a very perceptive researcher on human perception, and I'm glad to see this submission here. The lead example of faulty memory in the article is very interesting, as it points to the problem of belief-motivated recollection of "facts" that never happened. This is important for all of us, as the article notes. "Our lack of appreciation for the fallibility of our own memories can lead to much bigger problems than a misattributed quote. Memory failures that resemble Dr. Tyson’s mash-up of distinct experiences have led to false convictions, and even death sentences. Whose memories we believe and whose we disbelieve influence how we interpret controversial public events, as demonstrated most recently by the events in Ferguson, Mo." As I have written about this issue before here on Hacker News, the problem with human memory is that each read operation is also a write operation. We never go back to our memory of any event without rearranging our understanding of the event, often to the detriment of recalling verifiable facts. When in doubt, look it up. And be in doubt often when relying on your own memory.

api 12y ago
I also like that he picks on Tyson... not because I don't like Tyson, but because it's important to point out that these are universal human cognitive failings that can befall anyone regardless of their belief system. Nobody who has a human brain is immune to the issues with that organ. Too often "Skeptic movement" people tend to pick on religious believers, new agers, etc. as if obviously only those kinds of people fall prey to cognitive biases, illusions, logical fallacies, and other failure modes of the mind.
- DanBC 12y ago
  > Too often "Skeptic movement" people tend to pick on religious believers, new agers, etc. as if obviously only those kinds of people fall prey to cognitive biases, illusions, logical fallacies, and other failure modes of the mind. That was particularly annoying on the bad science forums. Especially since the whole point of the book was that people are subject to biases and well run research is the way to combat those biases.
WalterBright 12y ago

spiritplumber 12y ago

Why Our Memory Fails Us: Because we don't have badges. Cops have magically perfect memory, even to the point that their memory trumps dashcam video. https://www.techdirt.com/articles/20140405/17142626817/indiana-supreme-court-declares-officers-testimony-is-more-reliable-than-video-evidence.shtml https://www.techdirt.com/articles/20140405/17142626817/india...

dewarrn1 12y ago

False memories of the kind that Chabris and Simons discuss are ubiquitous in healthy people, and there are lots of interesting wrinkles in the literature describing these memory errors — here are two of my favorites. First, even individuals with extraordinary memory abilities are susceptible to false memory phenomena [0]. People with what's been called "highly superior autobiographical memory" (HSAM) can, for any given date in the last decade or two, tell you what they did, what they ate, and what the weather was like with extraordinary accuracy. However, when tested with standard laboratory tasks that induce false memories, they are just as vulnerable as typical individuals. Second, focal brain injury may reduce susceptibility to false memories [1]. Full disclosure, this is some of my own published work. We investigated the effects of damage to the ventromedial prefrontal cortex (vmPFC) on false recall in a standard laboratory test involving lists of conceptually similar words (the DRM task [2]). Patients with vmPFC damage showed reduced (but not abolished) false memories relative to healthy comparison participants. This is especially interesting to us because vmPFC appears to be necessary for decision-making in complex situations, particularly when prior knowledge must be integrated into ongoing behavior. We believe that deficits in decision-making accompanying vmPFC damage may stem from the same root cause the reductions in false memory that we observed. Regardless, we’re not recommending vmPFC resection for healthy people seeking more accurate memory (not yet, anyway). Although these are highlights, empirical work on false memories and other kinds of memory errors extends back more than a century. It’s a fascinating domain, and I very much enjoy working in it. More information on a variety of related topics can be found at my website [3] and in my published work (see Publications). [0] https://www.ncbi.nlm.nih.gov/pubmed/24248358 https://www.ncbi.nlm.nih.gov/pubmed/24248358 [1] https://www.ncbi.nlm.nih.gov/pubmed/24872571 https://www.ncbi.nlm.nih.gov/pubmed/24872571 [2] https://en.wikipedia.org/wiki/Deese%E2%80%93Roediger%E2%80%93McDermott_paradigm https://en.wikipedia.org/wiki/Deese%E2%80%93Roediger%E2%80%9... [3] http://david-e-warren.me http://david-e-warren.me

colmvp 12y ago

The article brings up Dr. Tyson. He actually commented on this article and linked to these Facebook posts: https://www.facebook.com/notes/neil-degrasse-tyson/email-exchange-with-the-federalist/10152354422065869 https://www.facebook.com/notes/neil-degrasse-tyson/email-exc... https://www.facebook.com/notes/neil-degrasse-tyson/partial-anatomy-of-my-public-talks/10152360009440869 https://www.facebook.com/notes/neil-degrasse-tyson/partial-a...

guelo 12y ago

Disappointingly, this article does not tell us why our memory fails us, only that it does.